1. Field of the Invention
The present invention relates to an image processing apparatus, an image processing method, and a storage medium.
2. Related Background Art
Researches for multiplexing another image in image information have been carried out well.
In recent years, as a digital watermark technique, a technique of multiplexing additional information such as an author name and permission/rejection of use in image information such as a photograph and a picture in such a manner that the additional information is difficult to visually recognize, and distributing the information through Internet or another network has been standardized.
On the other hand, in recent years, from a viewpoint of preferable image reproduction, an image output apparatus sometimes performs various correction processings with respect to an original image.
Examples of the correction processing include luminance correction, chroma emphasis, lightness correction, and the like. For example, the image is analyzed, and subjected to preferable chroma correction in the processing. Moreover, in a constitution for judging a natural image, and acting only on the judged natural image, a gradation image or another image can be subjected to a processing in which gradation is regarded as important.
Furthermore, when a digital camera image having a small number of pixels is printed as the original image, appearance of the image is sometimes inferior as compared with when the image photographed by the high-pixel digital camera is printed. To solve the problem, known is a method of using antialiasing techniques such as a bi-cubic technique and a nearest neighbor technique and enlarging an input image to an appropriate size to print the image.
Additionally, the image to be printed sometimes includes noise. For example, a color noise by CCD device dark current, a jpeg block noise during image compression, and the like are included. In subtractive color mixture, a gray line is often very unstable, and gradation is sometimes reversed in a periphery of the gray line. Here, as a problem, a color which is to be originally gray is replaced with the color in the periphery of the gray line by the color noise, and the reversed portion easily appears on the printed image. To solve the problem, as a method of reducing the color noise, a smoothing filter with a fixed size is used for dyestuff data. Furthermore, there is also a method of removing the block noise.
It is assumed that the watermark is used for a purpose preventing illicit counterfeit of paper money, stamp, securities, and the like, besides the purpose of protecting a copyright. For example, a special mark or a watermark is multiplexed beforehand in the paper money, stamp, securities, and the like, this mark is detected by an image output apparatus, information is regarded as image information of the paper money, stamp, securities, and the like, and a processing for discontinuing printing, issuing a warning, or intentionally coating the entire surface with black is performed.
First, when the watermark is detected in a system for printing image data of the paper money, stamp, and the like from a personal computer with image output apparatuses such as an ink jet printer connected to the personal computer, in an operating system (OS) of the personal computer, some application software (hereinafter referred to as “application”) is necessary for printing the image data. However, since an output from the application is emitted as a drawing command or a group of the commands, there is a problem that the output itself cannot be checked for presence/absence of the watermark. For example, in Windows operating system of Microsoft Co., Ltd., drawing is performed by the drawing command called GDI, but in some situation the presence/absence of the watermark cannot be distinguished unless the watermark is drawn and formed by separate GDI.
Moreover, when both functions of mark judgment and original image correction processings are introduced into the image output apparatuses such as the ink jet printer, the former processing for judging the presence of the mark is performed on the image data subjected to the latter various correction functions. In this case, since the corrected and modified image data is subjected to the judgment processing, there is a possibility that judgment precision is deteriorated, or that judgment becomes impossible.
Furthermore, in the aforementioned correction processing, there is a technique in which histogram creation means for creating a histogram based on pixel data of the original image is disposed, the pixel data accumulated from a predetermined pixel value and corresponding to a predetermined frequency is detected in the histogram, and color balance, contrast and chroma are corrected based on the detected pixel data.
On the other hand, in the image output apparatuses such as the ink jet printer, two sides of high-speed printing and high image quality printing are demanded, but it is actually remarkably difficult to establish both of these sides. Therefore, usually, a plurality of printing modes are presented, and a user can select the mode from user interfaces such as a printer driver in accordance with a purpose of that time.
Moreover, in these image output apparatuses, besides a so-called common sheet, a coated sheet, glossy sheet, glossy film, OHP sheet, and other recording media can be printed. In general with these apparatuses, the user also designates a kind of recording medium to be printed from the user interface, the printer driver or the like accordingly prepares printing data by an image processing, and an engine section of the image output apparatus performs printing based on the data. The image processing in accordance with the recording medium includes: correcting the color in accordance with a color reproduction range of each recording medium, or controlling an ink amount in accordance with an allowable ink amount per unit area of each recording medium for the ink jet printer.
In order to detect the watermark, there are methods of subjecting the image data to orthogonal conversion to detect the watermark in a frequency region, detecting a mark constituted of a plurality of concentric circles different in diameter in an actual space region, evaluating whether a pattern matches with a pre-registered pattern, and the like.
Among the various methods for detecting the watermark, since the method of detecting the watermark in the frequency region uses a heavy-load processing of the orthogonal conversion, many image memories and much processing time are necessary. Moreover, when the mark embedded in the actual space region is detected, the similar problem occurs. Because the pattern matching has to be performed in a broad area.
Particularly, it is assumed that the watermark is embedded in the paper money, stamp, securities, and the like, and such watermark is detected to detect the paper money, stamp, securities, and the like. In this case, no watermark is embedded in a large number of pieces of information to be subjected to a detection processing. However, in order to judge that no watermark is embedded, while detection conditions are changed, the detection processing is performed a plurality of times. This requires enormously much time.
Even in this situation, for expensive apparatuses such as a copying machine, the problem can be solved by disposing a huge memory or hardware for detection inside the apparatus. However, in the image output apparatuses used by the general user, such as the ink jet printer, in connection with cost, these processings have to be performed by the printer, or firmware or software in the printer. Therefore, the processing does not catch up with a printing processing speed of a printer engine, the printer stops, or another serious performance deterioration occurs.